·.·!'

. .

r. ,,

SUMMARIES OF MICHIGAN PAVEMENT ROUGHNESS 1961 Test Program

P. M. Schafer L. H. Holbrook

Prepared for Road Construction Division

Research Laboratory Division Office of Testing and Research Research Project R-47 F-15

Report No. R-395

Michigan State Highway Department John c. Mackie, Commissioner

Lansing, December 1962

SUMMARIES OF MICHIGAN PAVEMENT ROUGHNESS 1961 Test Program*

Synopsis

Results of field roughness measurements on newly constructed pavements are presented for the 1961 test year. Projects of rigid pavement (both two-lane and one-lane pours) and expressway-type flexible pavement are ranked according to pavement riding quality.

In addition, on the basis of new statistical analyses which are reported here, it is concluded that surface roughness of concrete pavement appears to be a function of project length. Longer pro­jects have been consistently and notably smoother than shorter projects in surveys of more than 300 construction projects during the last 11 years. The analysis thus indicates the importance of close supervision and quality control for smoothness on shorter construction projects.

Approximately 678 lane miles of pavement were surveyed in the 1961 roughness measuring program of the Research Laboratory Division. This is the greatest annual volume of lane mileage surveyed in the entire 11-year test program. Comprising this lane mileage were 485 lane miles of standard rigid pavement (two-lane pours), 158 lane miles of flexible pavement, and 32 lane miles of rigid pavement widening (one­lane pours). All surveys were conducted in the usual manner, with the same equipment and instrumentation used the previous test year.

Rigid Pavement Construction (Two-Lane Pours)

Individual rigid pavement projects constructed as standard two-lane pours, and their roughness values as determined in the 1961 test program, are listed in Table 1, grouped by year of construction and ranked within these years according to accumulated inches per mile of roughness by

* Throughout this report, the terms "construction year" and "test year" are specifically used to distinguish between the period of construction operations, and the time when measurements were conducted by the Research Laboratory. Further, the term "project mileage" refers to length given by the Contract Division, "roadway mileage" refers to length of two-lane pavement, and "lane mile" to length in terms of individual vehicle lanes.

z 0 ... u

~~

"'"' ~ ... -~

z 0 u

z 0

... u ~

a: ... "' z 0 u 0 ~

~

z 0

... u ~

a: ... " z 0

u

TABLE 1 ROUGHNESS DATA SUMMARY FOR RIGID PAVEMENT (TWO-LANE POURS)

" " ' "" "" ""

"'

projc•t'l

82053, c~n•, cun•

8201}~, Cl2U' B205:J, ClU' H2053, Cl2R'

8205:1, C6U' ~2053, CHU' 820!;3, Clf;Jt'

~110~, C7HN

EBBF 370\1, C21lN EBRF :1701;!, C4RN

BU 56023, C2U', C:IU

U ~2053, C5Jl'

~902!;, C5UN, C6HN

BF 29011, C6HN

BU 4ll3I, C·13UN', CHUN

IH' .;2!H2. cr,n BU 52042, C6U

m 41025, CIRN, C!H!N

nr 37012, c:mN UF :17014, C5RN

EBI H025, C51!N

Ill 73171, CSRN

BI ~1104, Cf>I!N

EllBP 76023, C3RN

BF 370H, ClltN

1\F Z901l, CSt!N IH' 29014, C!RN

l':IHW 370H, C71lN F.IHJF 1~0:13, CHIN, C7UN

l~tr!ct

'"

'"

"'

'"

t.un~th,

nti

;;,.J:!U

Type

~~It

(!lunl)

~~ fL (Dual) & Ufl

~~ ft

Uti ~-1 !o·om F<>r<l 1\d Pl 1-i:l), no>l"lh to not•!h t>!' lht• Ann Arlml' Trall

t:S :!I (>wuthlonund only) lrnm }",•nk1·ll Hoi (l'in• .\li!e [{!I), nurth tu Ct•uml Hll'<'l' Al't' (I"S ]o;J

1\'t•l~ht<-d A1·ithn>ell<" ~k:m f<>l' w.w C'l>n~(l\lt•Hun T<'Btc•t! in i!JIJI

I 9~ (<·ll~tl>u~n<l unlyj l!'o>m X\'1' !Ill, ,.,.~1 Itt t.··~!J ft w<·~t or Fkl<'ll<·~ nd

t!.70ll ~I fl (llYnll

n; ~7 l\U l!t•!ut· t~ut•Uo t'<IIIIIL't'Uon) ii'Ulll ~)!1!1 It ><<>Uih tlf llt•u•·fi..!<l lid, nurlhW<'>'I lu uh! l':; 2~, :~n<l (':; ~7

:I.:J:W

4.0:;r,

,;, 7~1

},000

1.!120

N.!l:m

4.111

5. ~90

:1. !)(10

:1.12:!

t·l rt (Utml), ~-1 rt & I>< It

UL'Im· from nurth "I lll:~nellard !!<!, mu·Ur nn<l ""''thwu><t \" ,~;,It nurlh ul Ut·o•l'fn·l<l !~!

~I~([ U<"lut• hvm Ell~w"l'lh ~~. Mldhmd, t':l><! tu ~ll<lh•nol·Bar t'" lino•

t·:; ~I lr<>m 11<-hun l!d, •~•rth lu 1-"<•llhl'lllhl Wi•·•• ~Ill"

ll<h

~~ fl \'S ~ [r<Uil llUI'th :III)II'<"IL'h <II ,\l;u·hin:!!' llrhii[L', n<H·(h (H

(IAmll ~. ~:w ml nu•·th ul nurtb llmH ul S!. lit'""'''

~~ ft L'S ~'i 11.-l<><' ln•m north uf 1\'a~hl"'\1<>!1 11<1 (t•:u<l 1>l llha~"h {IAmll & 11011b I<> !<IIUlh ul l.in<'<>ln l~l (c·;o~! of AlnH<I

~-1ft n; J:H lll'iuc !rum ~~th :;1, IV)uminA 01 II) nurth tu (Du:li) IJU!'Iun St, t:,·an•l ltaplds

~-l fl

~~II

(Du:LJ), I< ~I I

~-1 It (lAm !I

~·l It (!Junl)

~I II (IJulll)

21 j)

{!J\1:11)

2·1 j)

Hfl

(IJuaiJ

2-1 Jt (Dual)

~I l't (!Jual)

~-1 a (Dull!)

t·s ·11 - ~~ ~-' !rum I. a• mi W<'Sl uf M:~l'<(llt'L!t• W<·~l <'i!,V limit, •·a~l In t·:o~l uf lht• W<•S\ limit

ll"<>il':hl<·d Arltbn1etlc Mc•nn for H!!HI CnnMru~llun T<>stcclln l!Hll

I lim- \'S lH fi"Um Lcon:ml Sl, Grandlt:ll>lds, nuJ·th and w•·~t t" '"''~! uf t'S l:ll Bll (l>la!nl!dd ,,, . .,, t:l·nmlllal'hl><

1':; ~7 1\ll(north conm·clion) from ul<l liS ~7 In MI. l'luas;ml no•·th :•ml <>!IHllu t:s ~7 ll<'iuc, """ l'S ~7 Ht·luc from M ~o north tu mu1h <>f 1\nSL'I[UI<h ltd

lllfl - n; Hi 1'1~>111 I :!~I fl 11<>1'\h ul Casen<l<' H<l (\'S l6 Hll),

~uutll<':i><l h> n~>rthwo·~l •>I ~Hth Si (,\[ II)

I 7.1 lrunl ~:10 l't "'"'lh ul Tuwnlt>w 11<1, north to >«>Uth uf ll<oke>' Jl<l

l 9-l (L'asll>mond only) from tlw ,Jat•kson-1\'n~hlun:tw CL> line', uast tu NYC fll\

~I 7H J!chou fmm ~900ft ~molhw<·~t of Gmnt! HiV<'I' l~l,

northt'liSI to nnrlh<•m•l of ,\nn ,\rhor !Ill

t·s 27 lk•loc from mwth Hf l!usdou~l• Hd, north I<> north .,r J!urrick 1\cl

n; ~7 1\<'hlc from ~outh of J.lncoln Jl<l (<>l<l l'S ::7 L'<lst of Almn), ll<n1hwcst to OOl'lh of .\1 Hi

US 27 HL'iuc from n<>rth of li<'rrkk l!d ::wuth of Ci<ll'<', DL>r\h tu ol<l US ~7 north of Clan•

nl<lJ:rator, L.cvl'l !n./~l!lc ln•hcntor,

g'K "Mil"

101~

l!H

\:Ill li11

1!11

1-!11

l~li - !14ii

'"

lOii

II~

ll:l

Il-l liOI

ll!;

119 "" 121

1~2 7Uf>

m 4102·1, C5RN ~.OI!i 2-l It (Otml)

I !16- l'~ W from 9:10 n northwe~l of Tho>'napple I!Jvcr !)r 1~7 n:J;:;

4102·1, C7HN ~I fl (Dual)

cust to 900 n west of Whitn<·yvl!l<· !ltl

I uo - U:; IU f>'"m north ol ~tith Sl (~\ Jl). suuthea~l to ~~~

northw<•st L>l Thornn1>plc IU•·ct· Or

}'or \VIclcm!lg or '\hlrd-lanc • uonstruction nt thuse locations see noblu :J,

;;; Contract awarded to Cha~. J, Rogers Ine., Cooke Contracling Co,, & Jullon-K«llY Co. Pl Contt·act awarded toW. 1!, Knupp, Inc. nnoi w. J', McNally Co,

(4) ~::~:~~~"~~-~~~~ ~~ll~~:: g:;~:~.~~:n Cc"~. & Cha~. J. Ho~:ers Inc.

:~; Subcontract from llolloway Colll<trucUon CL>. ('I) Subconlr~cl from S. D. Solomon & :;ons (8

) Sub<o<>nlracl from A, LirulOOrg & Sons Inc, Contract (IW3t·ded to !lertc\-Deyo Co. & c. f:. Utterb;<Ck

-2-

Paving l'<>mr:oel<>r

Coukt• ConlraNI~ Cu. (ll

Bairloy & Lindley Inc.

U. J. llri!l"hlon

]lerLei-D~yo Co. (~)

W, 11. Kmop1>, hw. (2)

Cuuku Contn,.,lina;: Cu. (:I)

l.o~L'llt• Cun~tructiun Co. (·l)

Sa'lJ"Cnt Con~!t•ut•tion Cu. (S)

L. W. E<lbon Co.

L. W. Brumm

Sa'lJ"unl Cun~truction CL>, & Pierson Contrncllng Co. (G)

Denton Con~lruction t:o.

Sarg<•nt Cor.struetiun Co.(~)

Snrgont Construction Co. (J)

Pierson Cuntructing Co,

!lcnlon Con~truutiun Co.

Denton Construction Co. ('i)

L, A, !luvilis<ln

Sorgen! Construction Co.

l'iL'l'S<>II Contr.><•tillg Co. (GJ

! .. IV. f)Ji~on Cu.

TABLE 1 (Cont.) ROUGHNESS DATA SUMMARY FOR RIGID PAVEMENT (TWO-LANE POURS)

Rcughne~s

District Length, Jntegmtor, Level

Paving Contractor Project m(

Type Route and Project l..<xmtion Indlcntor, ln./Mile g's/Mile

'" 23061, C4RN, CGUN 5,281 24ft US 27 ftcloc from old US 27 211!!0 fl north of Five Points 130 511 L. W, Edison Co. CBRN Hwy, northco.st and north to existing US 27 - M 78

northeast of Charlotte

z '"' " 37013, C8RN 3, 370 2•1 ft US 27 Rcloc fr<>m B75 ft north of Deerfield Rd, north to "' '" Hertel-Deyo co.

(Dual) M 20 0

B< 41025, C3RN* 2.50() 24ft I 96 - US 10 from Luonnrd St, Grand Rapids, southeast 134 ,, L, W, Edison Co.

(Dual) to 1321 ft north of Cascade Rd [US 16 BTl.)

EBBF 11081, CUR ., 2, 244 "" I 94 DL from ~·air St, Benton Harbor, east to I 94 t:\4 611 Carl Goodwin & Sons, Inc. "' ~ (Dual),

48ft, &

u 6(} ft EBBF 76023, C2HN 6,590 24ft M n R<Jloc fr<>m 850 ft cast of M 47, norlhc"st to 135 505 C. 1'. Heploglc Co,

{Dual) 2000 ft aonthwest of Grand River /ld

~ EBBF 58033, C2RN " :1.10~ 2~ ft US 2:1 Rdo<: (northbound only) from lH ft n<>rth of 137 732 S. J. Groves & Sons Co.

~ Allll Ar~or Jut, north to Monroc-Washhmnw Co line cost of ~Ulan

BU 41131, C55UN' 1.5:17 2-1 fL · US 131 I \doc from Burton Sl, GraU<l Rupids, north to HO '"' Hcrtci-Doyo Co. ~ (Dunt) Franklin St (M 21 BR)

~ BM 37015, ClRN 1.090 2-1ft US 27 B[{ (south connection) [rom Herrick Rd, northwest "' 76•1 Denton Construction Co. ,,,

(Uuul) to old US 27 south of C lnrc

z B( 70063, CIORN G.25-! 2-Jft !190- US IG from cast of lGth Ave {cnst of Marno), t-J2 '"" Carl Goodwin & Sons, fnc,

B< <11026, C3RN (Dual) """t to mu;t of Bl"lstol Rd (northwest of Grand Rrtplds)

0 EJHlF 58033, CIRN " 5.~73 2-1ft US 2;! lluloc (northbound only) from 1050 It north of M 50, 143 "" s. J. Gt-oves & Sons Co.

(l>tml) north tu 1-Ll n north of Ann Arbor JlR u

B( 73111, C9RN ~.51$ 2-1ft 1 n (~outhhcund only) from :mar, ft north of In; BL '" "" Cooke Contracting Co. (southeast of S<t!l"lnuw), north to ~~89 ft north of M -lfl intcrchnnge

B( 73171, C-lUN, C6HN J.093 2-1ft 1 7,; from Birch nun R<l, north to 2M fl north of '" "" Sargent Construction Co. (Du~L) Town/inc lid

~ EF 6~052, C!oiH* '" 0,56-1 2~ ft US 24 (aolllhbound only) from Haskell st, north to 34~ ft 149 ~:12 Cook<• Contracting Co, north of Hayes st.

~

''" -tll:n, cn~UN' 0,624 2-1 tl US t:ll R<>loc from 1-"rnnklin Sl, Grand Rapids (M 21 BR), '" 1073 L. W. Edison Co. (10)

(Dioul) north to Goodrich St

ou 24011, C5U 0,587 21 n US 3l!Wioo from near Creak brttlgc, Petoskey, 2-14 1438 Hodgkiss & Douma, Inc. (Dual) & northaast tc old US :n (Bay View st) 2-1 (t

Weighted Arithmetic Mean for 1961 Construction "" '"' Tested In 1961

WElGl!TED ARITHMETIC MEAN FOR 1961 ROUGHNESS TEST YEAR 13-1 "" For widening or "third-Jane" construction at these locntions see Table 3.

"' Subcontract from A. Lindberg & Sons Inc.

'"' Contract awarded to Hertel-Deyo Co. & c. E. Utterback

'"' Contract awarded to Canonle ConstrueUcn Co. & Carl Goodwin & Sons, tnc. (10) Subcontract from L. W. Lamb Co.

-3-

Integrator measurements; where two or more projects have the same Integrator count, they are ranked by Level Indicator count. During the 11 years of the roughness surveys, these Integrator values have ranged from a low (smooth) of 93 to a high (rough) of 282. This year the range was from 103 to 244.

On the basis of riding quality, the Laboratory classifies projects in three Integrator-count categories: "good" (0 to 130 accumulated in. per mi roughness), "average" (131 to 174), and "poor" (175 or more). Table 2 shows that since 1961, with a total of 323 rigid pavement projects tested, 44, 45, and 11 percent of this total have been good, average, and poor, respectively. In the 1961 test year, 38, 49, and 13 percent of the 37 projects measured were good, average, and poor, respectively.

TABLE 2 ELEVEN-YEAR ROUGHNESS SUMMARY FOR RIGID PAVEMENT

(TWO- LANE POURS)

TI:HI Ycnr 19fol-1961

Totil Projoct• " " '" " " " " "' " :15 " :l23

Pcr~'<>nllli!!!!! " '" 2U :IG " " H "" '" '" " O-I:Kl tn./mi

Pcrecnt Awrm•c :1/; "" <i1 " "' <i2 :16 '" ,, H " " ial-17~ in./ml

Percent~ " 27 Fi ~a " " u )70 or more ln.;ml

Weighted 142 lf>2 '" '" JjX HI 1~6 "" 12<1 "' "' 1~1

Artthmt•Uc Mean

Project ~m.,.....,• -16.:127 U\.57~ aa,7at -ll.\ITL ~~.<.90 82,H:I tGr,.o•a 1:1-LO-l~ \US.~92 IM,333 I.~J.OIJ IH0.529

Laoo MHO"l!e" \00.5H 16:1.310 2a:1,n7 91,022 1-10,:;7-1 230,:l(IU :;:;s.1~1 -1(;1.52{1 6-10.962 55-1.350 4a4,0L2 3664,801

A• nlven In Controot Division monthly "llop<>rt of Award•" .. Toto! n>.lleoge o( 11- o:r 12-n wide Innes

The weighted arithmetic mean for roughness of all projects tested this year increased 17 in. per mi over the mean for the preceding test year, as shown in Fig. 1. Two short projects with extremely high Inte­grator counts were significantly involved in this rise in the weighted arithmetic mean, both being less than 0. 7 mi in length. The roughest pavement this year was 0. 6 mi long, and had an Integrator count of 244 in. per mi; this is the roughest project encountered among newly constructed pavements since 1953, and the third roughest project measured in the whole 11-year test program. The extreme degree to which this project alone departed from the norm helped raise the mean for this test year.

Partly because of these short, rough projects, the "general trend toward improvement in overall pavement smoothness, " cited in the 1960 roughness summaries (Research Report No. 366), did not continue in the

-4-

TOTAL

100,-----------------------------------1-

"c000" RIDING QUALITY 0 -130 IN./M I ROUGHNESS

p 7~!-------------------------------------------.~---------l u ~0 }------------------------------------1- !l--~·~:~:::!1-: ---r. a. ~ 2~ _ ],1------------1 r.»"·

07-~m:L: __ ~~ .. ~·--~[J:~::::L-~~~·i~~:--__ ~~:~~~~-~rn·-:~:::-:L---~~~-ll~~~~~--~~

p t-U zw ,...., uO a: a: ..... a...._

0

I I

1001-1------------------------------------ ':.0.VERAGE" RIDING QUALITY

131-174 IN./MI ROUGHNESS

75~---------------------------1

w: m ::::: )\ :::::: '::':l

!IO !------~=~=~~-- ::::: ::it- [1[1[1------------------------.. .. "" .. ·.---1

.. -~~ 1~ 1--J~ 1--11111--J--1--J~--0==~=-~i -~i-01

I

I

IOOrl ________________________________ __

175 "POOR" RIOING QUALITY I

OR MORE IN./MI ROUGHNESS

751--------------------------------------------------------~

PROJECTS 17 22 40 17 22 21 33

TEST YEAR 1951 1952 1953 1954 1955 1956 1957

160r-------------------------------------

34 45 35 :07

1958 1959 1960 1961

WEIGHTED ARITHMETIC MEAN (ALL PROJECTS)

Fig-ure 1. Annual roughness comparison for rig-id pavement projects.

-5-

1961 test year. The percentages of all projects tested in the "poor" and "average" categories increased, while the percentage of "good" projects decreased.

Correlation of Roughness Measuring Instruments

The Laboratory's rouglmess measuring instruments are the Integrator (used since 1951) and the Acceleration Level Indicator (added in 1959 for supplementary and more extensive measurements, and intended for eventual

1600

/ / y /

1400

STAI\OARD ERROR OF ESTIMATE:: i68 .5 ''"""'"~~~~~/ "' " l 1200

~ w • ~ .• a.' I 000 0

5 i5 ~

" 80

"' > "' " z 0

~ ~ 60

"' " ~

20

0

0

0

0 90

.~/ ' I--•/ "'--

/ I ·~--:' ~y = 6,68X-197.6 ;:;,; I'= 0.934

·~i I ;;)';. 0,~ ~o f1J ~_.o

1'". 0 0 /o 0 oo% /no /(Sif..~V. • •

/oO J o?fo-;O (D 0 00 )-' tP

/ (9 o]o;e J o :/o 0

o"o/ . /.

/

110 130 150 170 190 210 230

INTEGRATOR, INCHES. PER MILE CX)

Figure 2. Comparison of rigid pavement roughness measurements by Acceleration Level Indicator and Integrator methods.

250

replacement of the Integrator as the primary rougnness measuring instru­ment). Correlation between the two instruments in their three years of combined use, as shown in Fig. 2, gives a standard error of estimate of

-6-

_:: 68.5 g's per mile (Acceleration Level Indicator value) and a correlation coefficient of 0. 934, indicating that the two instrumental systems are producing consistent and reliable results. This agreement between the instruments has resulted in immediate discovery and correction of occa-· sional minor errors or irregularities in field measurements.

Relationship of Smoothness and Project Length

On the basis of logic and theirownexperience, most highway engineers would probably agree With the proposition that the shorter a rigid pave­ment project is, the greater is its likelihood of having a rough surface. It is undoubtedly true that as construction progresses at a given site, the men and machines generally tend to improve in efficiency in proportion to the extent of their activities. Thus, other things being equal, later pavement is often likely to be better pavement.

To the authors' knowledge, no statistical analysis has even been attempted specifically to demonstrate this relationship between length and smoothness. In view of the marked influence that a few short, rough projects exerted this year on Michigan's long-range figures for pavement roughness, such an analysis was made, including data from 320 projects* measured since 1951, totalling 1892 roadway miles, and varying in length from 22. 9 to only 0. 1 roadway miles.

This statistical analysis took the following forms and did show an association between smoothness and length for these projects, as follows:

1. Scatter Plot. Roughness in accumulated inches per mile (Inte­grator count) was plotted against project length in roadway miles as shown in Fig. 3, in which the good, average, and poor categories of riding quality are delineated. Using 6 roadway miles as an arbitrary demarca­tion between "longer" and "shorter" projects, all longer pavements are average or good in riding quality, while shorter pavements appear only in the poor and average categories of riding quality.

2. "t" Test. Mean lengths for projects in the good and poor cate­gories of riding quality were found to be 7. 620 and 1. 867 roadway miles,

* Three other projects were excluded from the total323 surveyed, because of unusual design or construction characteristics, such as short or irre­gular slab length, etc.

-7-

respectively. To test the difference between these means, all data from the two categories were subjected to a t-test, a statistical technique used to assess the reliability of differences in such sample statistics as the mean, variance, standard deviation, etc. A definite difference between these data is evident in the t of 3. 37 3 obtained for 120 degrees of free­dom. The pertinent distributions are illustrated in Fig. 4. Further, Fig. 5 illustrates that a similar difference also can be obtained between the roughness distributions of shorter and longer projects, again using 6 roadway miles as an arbitrary demarcation between length types .

... •w ... ... "0

ao

•ro ... ... <00 · . ...

'" ..... """ 0 ..... :" ....

•• , • • • • • 'b .. • 0

•• ·: • • : 0 •• • 0 ••

.. '

v "'¥+ 131.~ f" 0.373

. 0 0

--- -.~-- ~-- --- -1~

I

·I :.~J--: 0 0

0 0

. . . . ----~

0

: too ' z

' 0 . ' ~10111213141"'""1810

LEN<lTH IN 1\Q~OWAV ..... .U

21 •s 20_..

Figure 3. Scatter plot of pavement roughness versus roadway length for 320 Michigan projects, with correlation curve.

3. Correlation. Carrying the investigation further, a correlation analysis was performed, producing a correlation index of 0. 373 (signi­ficant at a confidence level better than 0. 01) using the hyperbolic curve

y = 13.91 + 131.5 X

(1)

The curve superimposed on the scatter plot in Fig. 3 was taken from the 320 projects; curvilinear correlation was used because, as may be observed, the relationship between roughness and length obviously is non-linear.

-8-

4. Means vs. Length. In an attempt to reduce the data given in Fig. 3, the roughness means of successive 1-mile increments of roadway mileage were plotted as shown in Fig. 6, These means follow the same general trend toward geometrical decrease in roughness as project roadway mile­age increases.

~ z w

45

40

35

30

u 25 ~ w ~

> u z 20 w ~ 0 w a: ~

10

5

0

J ~ II I

i I I I

I ' I I I J I I ' I I I I

I I I I POOR RIDING QUALITY ! 175 IN PER MILE OR MOR£) I IV I

I I I I I I I I

I I I I I GOOD RIDING QUALITY < 130 IN PER MILE OR LESSJ I I ' ' I I ! A

~ I X' II """'- ~ r+-

I \: I

\ :r v I ~ ~ I

~ I

I

0 2 4 6 8 ~ 12 ~ ~ 18 ~ u ~ PROJECT LENGTH 1 ROADWAY MILES

Fignre 4. Frequency distributions of project lengths for "good" and "poor" categories of riding quality.

5, Standard Deviations vs. Length. Finally, perhaps the most signi­ficant of the various statistical determinations made, in its indication of the role and importance of quality control in construction operations, is

-9-

the plot of the relationship between standard deviation of roughness and roadway mileage. This is also illustrated in Fig. 6, where the standard deviations of successive 1-mile increments are plotted. A definite de­crease in roughness standard deviation appears as roadway mileage increases, indicating that these longer projects are characterized by less variation in roughness. Roughness data clearly are statistically more predictable, more reliable, and more consistent as project mileage increases. The geometric relationship between roughness standard deviations and roadway mileage, like that between roughness means and roadway mileage, is one of decreasing roughness as roadway mileage increases, or more precisely, it is a decay function as indicated in Equation 1.

28

24

20

c ZIO w ~ w a

> ~ 12 w 0 0 w « " •

•

0

I I I I

l I I I I I

I

I

!II I I

v

_A I I I I I \VI" LONG f 6.000-24.000MILESl

I I I I I ~ 1---.

P'-, \ ,; ' ' '\ ' \ \

I I

\ \

-..J \ \ ' \ ' '\....-r\

~SHORTt 0 -5.999 MILES I

I I 1\-', -r--~~ r--,_

90 100 no 120 130 140 150 reo 110 teo 190 200 210 22o 230 240 250 2eo 210 2 eo ROUGHNESS, INCHES PER MILE

Figure 5. Frequency distributions of pavement roughness for "long" and "short" projects.

To eliminate the possibility that some combination of changing con­struction methods, gradual modification of test methods, and the long-term increase that has occurred in average project length, might have produced

-10-

the trend in question, roughness and roadway length since 1951 were re­analyzed by successive two-year periods. Approximately the same rela­tionship was obtained in each of these five successive two-year periods, so that construction methods, survey methods, and increasing average length do not account for the trend.

4B

44

40

z Q 20

~ >

" " 0

0

~ 12

z ~ . "

4

0

\

\

0

I I

I \

\'-\ \

~STANDARD \ ,/',..

\; v - ~ ', !'v ' ,_ r-,

\ \

\

DEVIATION

I r- --- -- -- -- --

I

'-_\MEA~ \ I \1

-- --

I

I

I

I

I

I

70

eo

50

" 40~

~ i5 ~

30~ ...

20

" >

I •o 2 3 4 5 6 7 6 9 10 II 12 13 14 15 16 17 16

PROJECT LENGTH, ROADWAY MILES

Figure 6. Roughness standard deviations and roughness means versus project length in roadway miles as computed for successive 1-mi increments.

Thus it was concluded that by various statistical methods, the rela­tionships between project length and pavement roughness were clear enough, for these projects, to indicate a trend toward increasing smooth­ness with increasing length. Because the t-test results will not occur by chance factors alone as often as one time in a thousand, a conclusion that roughness is a function of project length definitely is warranted.

Given this quantity of data and this trend, it is apparent that close supervision and quality control are particularly important for smoothness on shorter construction projects.

-11-

Rigid Pavement Construction (One-Lane Pours)

In ;:tddition to the usual surveys of roughness on newly constructed standard rigid pavements (two-lane pours), the 1961 measurements in­cluded 10 rigid pavement widening projects (one-lane pours), with the results shown in Table 3 and Fig. 7.

The testing and reporting procedures used for these projects are the same as those for standard rigid pavements. However, due to somewhat different construction procedures required in pours of one lane-width, the range of roughness values varies somewhat from that for standard rigid pavements. For this reason widening projects are reported and tabulated separately from standard rigid construction. Table 4 sum­marizes test data obtained during the four years in which this type of construction has been under study.

Flexible Pavement Construction

Seven projects built to Interstate expressway standards were surveyed in 1961. The accumulated inches per mile figures presented in Table 5 are the result of measuring runs in separate wheel tracks in the traffic and passing lanes.

As in the case of rigid pavement widening, this type of roughness measurement represents a supplement and extension of the Department's roughness program, and will be included in all future annual reports as construction warrants. Normally, only flexible pavements of expressway quality will be included in future surveys, although other bituminous pro­jects may be measured when special roughness conditions make this desirable.

-12-

TABLE 3 ROUGHNESS DATA SUMMARY FOR RIGID PAVEMENT WIDENING

lONE-LANE POURS)

Proje"t District

z 0 ~~o5:~. C~R, C9H

~ .,u ~" ~«

1- U 82052, Cl3U II) U 82053, Cl U "

Length,

mi

1.8~6

~.498

Type

Uft

12ft

12(1

Hoot<• uml Project \.<>cation

US 24 (median si<Jc, southbound only) from Fenkell IW (Five M!lc JW), north to Grand RIVer Ave (US tG)

US 24 (median sides, lxlth roadways) from Ford ll<l (M 153), north to the Ann Arbor Trull, nod US 24 (mcdhn side, northbound only) north of the Ann Arbor Trail

us H (outer side~, both roadways) from Oxford nd, north to Ford R<l {M 15:!)

Roughness

Integrator, level

Jn./Milc tmlicntor,

'• HG 79·1

"" 78·1

"' 993

Paving Contractor

H. J, Drighton

Cook" Contracting Co. {l)

Balrley & Lindley Inc.

Z F !12053, CHR 3~----------J_ __ J_ __ j_ __ _L ____________________ _L ____ L_ __ L-----------~

z 0 ~ u

0:> ~ .. ~ ..

"' z

U !1205;1, C5R

BU 5G02:1, C~U

BU 411;11, C-t;Jt;N, CHUN

0,957

1.000

I~ ft

12ft

12ft

W~!gbtcd Arllhmc\l~ Mean for 1!150 Construction

US 2~ (mcdlun ~ides, both ro;t<lways) from Belton Rd, north to f'enkell Rd (f'!Vc Mllc l!d)

M 20 ltcloc (outor sldos, both rolltlways) from Ellsworth St, ~tldhmd, east to Washington HI

US PI lluloc (m~'lilnn ~Ides, hoth ro~dways) from 2Hth ~1., Wyoming (M U), north to Bur!on St, Grall<l llni>lds

"" ""' "' "" Cooke Contracting Co. (2)

'" "' w. II. Knapp, Inc. (3)

"' 1119 L. W. Edison Co.

er-------------~---L--~----~-----------------------L----~---L-------------4

z 0

~

u

" a: ~

"' z 0 u

m .;Io25, canN 2.~00

BU Hl31, C55UN 1.537

0.56~

UU Ht:ll, C62UN 0.624

I~ It

12ft

12 It

12ft

W<•l~o~htc<l Arithmetic Mu:m for !DUo Constn•ctlon

1 96 - t:s 11; (mL~ltnn sides, h<>lh roadways) from i-"""''rd f.t, Grand Ra.,Jds, Houthe:tat to 1~21 It north of Cascade Rd (t'S JG BR)

US 1:H ll<'loc (median shi<'ll, both ro,.<.lwny~) from Burtun St, Grand lln.phl5, north 1.o Pnmklln St (M Zl BR)

US 2~ (mc<.liun side, southbound unly) from ll•t~kcll St, n~>rth to 3~7 ft north of U:•rc~ Sl

US t:H !!doc jmL'<It:m sides, both ro11dway~) from franklin St, Gl':tnd IU1pld~ (M 21 Bit), north to Gt.><><lrich St

Weighted ,\rlthmcllc Mean for 1961 Construction

IVI::IGHTEO AIIITIIMETIC MI::AN ~·on 195!1-61 CONSTRUCTION

(l) Contract awarded 1.o Cha~, J. Rogers Inc, , Cooke contracting Co,, & Jutton-Kclley c-o. :~: Contract awarded to Cooke Conll'ac\lng Co. & Chns, J. Rogers Inc,

Contract awarded toW. H. Knupl' & W, F. McNally (4) SUbcuntr;~ct from L. W, Lamb Co,

• All cunst ruction Ia "third-lane" widening to projects reported tn Table l.

-13-

155 865

162 - "'" L. W, Edison Co.

163 ... Hertci-Deyo Co .

173 832 Cooke Contrnctlng Co.

'"' ll60 L, W. Edison Ca. (4)

"" '" '" ""

TOTAL

~ t-U zw W"> uo a: a: wa. a. ....

0

12 t-U zw ..,.., uo a: a: UJQ. a..._

0

100 .------· I-~

11C00011 RIDING QUALITY 0-130 11'1/MI ROUGHNESS

7~~---------~

~Of-------------1

2~--f:f:l-------------~

:::\:\ m oi,L-~~----~~------~

I

1oo,1-----1 ':a,VERAG£ 11 RIDING QUALITY

131-174 IN./MI ROUGHNESS

7~f--~----------i

~Of-----

1 I I

10o 1r------1 "POOR" RIDING QUALITY 17~ OR MORE IN./MI ROUGHNESS

~

75~---------------------~

PROJECTS ·3 2 5 10

TEST YEAR 19~8 19~9 1960 1961

0 UJ t- w <Cil.J .JW-:>I: ::I! ::!!Ua: ,z.., u-a. u <

200

175

150

125

100

WEIGHTED ARITHMETIC MEAN (ALL PROJECTS)

Figure 7. Annual roughness comparison for rigid pavement widening projects.

-14-

TABLE 4 FOUR-YEAR ROUGHNESS SUMMARY FOR RIGID PAVEMENT WIDENING

(ONE-LANE POURS)

Test Year 1958 1959 1960 1961 1958-19()1

Total Projects 3 2 5 10 20

Percent Good 33.3 0 20 0 10 0-130 in. /mi

Percent Avera{{£ 33.3 50 60 70 GO 131-17..t in.,mi

Percent Poor 33.3 50 20 :JO 30 175 or more in./mi

Weighted 130 19-l l.JO 161 151 Arithmetic :\lean

Project :\Iilcage * 6.-103 3.092 l:L 925 17.70-l -11.12-l

Lane ~Iilcagc ** 10.533 3. 872 24.152 31.679 70.236

' .As given in Contract Division monthly "Report of Awards"

'* Total mileage of 11- or 12-ft wide lanes

-15-

I

~ I

TABLE 5 ROUGHNESS DATA SUMMARY FOR FLEXIBLE PAVEMENT

Leng·th, Project District mi

m (ill5~, C5RN :) il.:!87

EBF 7201~, ClHN 4 9.G-!3

EBI

BI

EB! EBI

EBI

BI

11015, CIOHN, Cl-!UN 7 7. 87;;

11015, C8RN 7 G.7J9

7006:-!, CORN ' 3.020 7006.±, ClRN

70063, CBRN ' -!. ::580

70064, C3RN 5 2.053

(l) Subcontract from Holloway Construction Co. (2) (3) (4)

Subcontract from Canonie Construction Co. Subcontract from Maclean Construction Co. Subcontract from S. J. Groves & Sons Co.

Type

2-l: ft (Dun!)

24ft

(Dual)

:!.-:1: ft (Dual)

:!.-! ft (Dun!)

:!~ ft (Dun!)

~4 [t

(Dual) 24ft (Dun!)

Roughm.!s::;

ntcgrator LCvcl Route :md Project Location Indicator, In. /i'o.Iile

g's/i\Iilc

I 196 - US lG from 300 ft southc:u:>t of US :H, southeast 91 :39:3

to the ivluskegon-Ottawn Co line

t:S :!.7 Rcloc from 9. !Ill mi south of :\1 :30, north to 90 :!.60

0. -!G:!. mi north of ;\I 5f>

I 9·1 from Snow Hd, north to Hidgc Rd 101 -!67

I 94 from 29·!7 ft southWl':.;t of Easy Hd, north to 106 -HiG Snow Hd

I 19() - [S Hi from :li'lO ft northwe::;t uf StaLL' Hd, 117 .±0-! southeast and east to Fitzgerald St

-I 19G - US 16 from Fitzp;crald St, t•ast to 72nd ..\\'t' 119 ;)!)1

I 19U - L'S 16 from l\Iuskegon-Ottawa Co line, southeast 1~-i ;)02

to 2730 ft northwest of State Hd

Weighted Arithmetic !\lean for 1901 Construction 104 ~1.j

- -

Paving Contractor

Reith-Hiley Cont;truction Co.

Ann Arbor Construction Co. & Lake & HO\Vell Co.

Hcith-Riley Construction Co. & Globe Construction Co. (1)

Globe Constnwtion Co. (:n

I

Paul C. :\Iiller(:J)

7\lichigan Colprovia Co.("!)

Paul C. :\Iiller